Re: Use VFPv3-D16 FPU for ARM builds
Here's the bug in LP and the trac issue it links to
the D16 part was Debian/Ubuntu specific, IIRC we define hardfloat in that particular variant (16 double registers) or we had a different naming for some reason.
I do not recognize the error message in your log though, it was something more telling in the bug I chased down back then.
Unfortunately I do not know about the current status of ARM in GHC. I know Colin Watson worked on ARM64 but that may not translate to ARM.